Subjective Knowledge Representation for Intuitive Human-Robot Interaction Based on Mental Image Directed Semantic Theory (pp. 75-124)
Authors: (Masao Yokota, Department of System Management, Fukuoka Institute of Technology, Japan)
Abstract: This chapter gives an innovative knowledge representation system in order for integrated management of knowledge subjective to cognitive mechanisms of intelligent entities such as humans and robots based on a mental image model and its description language. There were proposed several kinds of knowledge representation methods or technologies and thereby a considerable number of knowledge representation languages (KRLs) were invented for formal ontology of the world, namely, formalization of (concerned portions of) the world with such an intention so as to facilitate inferences involved. However, almost all such ontological commitments about the world were quite objective, namely, without any reflection of human cognitive processes or products although ordinary people live greater part of their casual life based on their own subjective knowledge of the world acquired through their inborn sensory systems. Actually, according to the author’s long-termed study, it is very clear that people’s concepts about the world expressed in natural language are very much affected by their cognitive processes, which is very natural considering that they are very cognitive products in humans. For example, people often perceive continuous forms among separately located objects so called spatial Gestalts in the field of psychology and refer to them by such an expression as ´Four buildings are located in square. This fact means as well that knowledge representation without reflecting human cognition is not very useful for interaction between ordinary people and robots especially by natural language, so called, intuitive human-robot interaction. The author has already proposed a KRL named Lmd (Mental image description language) in his original semantic theory, Mental Image Directed Semantic Theory (MIDST). The KRL Lmd, quite distinguished from others, was invented in order to model mental images subjective to humans and he has found that the word concepts concerning space can be formalized systematically in Lmd by introducing spatial Gestalts as virtual objects. The following sections describe the outline of MIDST and the details of Lmd and its application to knowledge representation in order to facilitate intuitive human-robot interaction.
Open Access item.
Click below PDF icon for free download.
This is an Open Access item. Click above PDF icon for free download.